[
https://issues.apache.org/jira/browse/ARROW-11840?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ben Kietzman resolved ARROW-11840.
----------------------------------
Resolution: Fixed
> [C++][Compute] Support merging GroupByState for multithreaded aggregation
> -------------------------------------------------------------------------
>
> Key: ARROW-11840
> URL: https://issues.apache.org/jira/browse/ARROW-11840
> Project: Apache Arrow
> Issue Type: Improvement
> Components: C++
> Affects Versions: 3.0.0
> Reporter: Ben Kietzman
> Priority: Major
> Labels: compute
>
> ARROW-11591 adds support for grouped aggregation, but defers merging (which
> is non-trivial and unnecessary for single threaded aggregation). Eventually
> it will be required to support merging, however: when aggregating in a
> multithreaded dataset scan, each thread's results will need to be combined
> after the scan is completed.
> Note that currently {{ScalarAggExecutor::Consume}} assumes that merging
> aggregations is not costly (true for small aggregation state as with "mean",
> but false for "group_by"), and invokes {{ScalarAggregateKernel::merge}} for
> each input batch. ARROW-11591 introduces "group_by" as a special case which
> will not be merged for each input batch but Ideally this assumption would not
> be made for any kernel. When removing it, be sure that merging other
> aggregates continues to be tested.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
